{"id":"c93a8a41-e229-4f88-b0a0-91cf8d42871d","arxiv_id":"2501.07520","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"3D particle tracking in hydrodynamic disk simulations shows that small dust can cross a giant planet's gap in both directions via meridional gas advection, with the crossing fraction depending on grain size, disk viscosity, and planet mass.","lead":"This paper simulates a protoplanetary disk with an embedded giant planet and tracks individual dust grains in 3D, finding that small grains can be carried across the planet's gap by gas flowing toward the planet. This matters for explaining how meteorite material moved between the inner and outer solar nebula while Jupiter was growing.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 10–30% crossing fractions depend on the Schmidt-number-unity diffusion law in Eq.","rationale":"The reader identified the Schmidt-number-unity diffusion law as the weakest assumption, and my read agrees. The central mechanism—advective carry-over of well-coupled solids in 3D meridional flows—is not in question: the trajectory in Figure 6 is a direct demonstration, and it is consistent with prior 3D gas-flow studies. What is load-bearing for the paper's headline percentages is the supply of small grains to the high-altitude inflow. That supply is controlled by Eq. (6) and the stochastic vertical transport of Eq. (3), together with the unstated initial vertical placement of particles. A modest increase in Sc or reduction in vertical diffusivity reduces the population at 1–3 scale heights by an order of magnitude, which would directly shrink the reported 10–30% crossing fractions. I therefore recommend keeping the existing CONDITIONAL verdict: the qualitative claim is supported, but the quantitative mixing fractions should not be used for meteorite-record interpretations until the diffusion law and initial vertical distribution are tested. No external code is provided, so a controlled rerun with modified diffusion parameters is the most direct check.","tokens_in":25905,"tokens_out":12212,"duration_ms":133282,"concrete_test":"Repeat the Section 3.2 particle integrations for the 100 and 300 M_⊕, α = 10^\\u22123 cases with D_g = ν/Sc for Sc = 3 and Sc = 10, and separately with anisotropic diffusion (D_r = ν, D_z = ν/10), using the same FARGO3D snapshots, timestep criteria, and 1000-particle sampling. Also rerun the fiducial Sc = 1 case with particles initialized from the gas-density vertical distribution rather than the currently unstated default. If the 1 Myr fraction of 10 and 100 µm grains crossing 5.2 au remains above a few percent, the mechanism is robust; if it collapses below about 1%, the reported 10–30% values are an artifact of the Sc = 1 and initial-condition choices.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing assumption is that small grains are able to populate the high-altitude, inward-flowing gas that carries them across the gap. The particle diffusivity is set to D = D_g/(1+St^2) with D_g = ν = α c_s H, i.e., Schmidt number unity (Eq. 6), and vertical transport is modeled as a uniform random walk in position (Eq. 3). For 10–100 µm grains at 5.2 au, St ~ 10^\\u22125–10^\\u22124, so the equilibrium dust scale height is approximately H_d/H ≈ sqrt(α/[Sc(α+St)]). With Sc = 1 this is close to H, meaning a substantial population exists at z ~ H where the meridional inflow onto the planet is strong. With Sc = 3–10, H_d/H drops to roughly 0.6–0.3, and the number of grains at 1–3 scale heights falls by about an order of magnitude or more. Because the crossing trajectories shown in Figure 6 and the statistics in Figures 7–8 depend on grains first diffusing off the midplane, the quoted 10–30% crossing fractions are not robust unless the diffusion law is independently constrained. The paper provides no check of Sc or of the initial vertical distribution of particles, and the 1 Myr integration time amplifies the role of D. The qualitative central claim—that well-coupled solids can be carried across the gap by 3D meridional advection—is physically plausible and supported by the trajectory in Figure 6, so the concern is primarily about the numerical percentages, not the existence of the mechanism.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper combines 3D FARGO3D hydrodynamic simulations of a protoplanetary disk containing an embedded giant planet (0-300 Earth masses at 5.2 au) with a Monte Carlo post-processing particle tracking method to study the transport of dust grains from 10 micron to 1 cm. The central claim is that for planets above the pebble isolation mass, small, well-coupled grains can be entrained in the 3D meridional gas advection onto the planet and cross the gap from the outer disk to the inner disk (and, for 10 micron grains, also outward). The paper reports up to ~30% crossing fractions for grains smaller than 100 micron over 1 Myr, identifies multiple outer-disk dust pileups in low-viscosity disks, and interprets the results in terms of the NC-CC meteorite isotopic dichotomy and a possible tertiary reservoir. Section 4.4 acknowledges limitations: constant gas structure, fixed planet orbit, finite particle number, and no grain growth or fragmentation.","tokens_in":26226,"tokens_out":8913,"duration_ms":81018,"significance":"If the quantitative results hold, the paper provides a concrete 3D mechanism for exchanging small solids across a giant planet's gap, with direct implications for the interpretation of isotopic reservoirs in the solar nebula and for the filtering of dust in exoplanetary disks. The methodological contribution is noteworthy: coupling a high-resolution 3D gas simulation with a particle tracker that adjusts the Stokes number as particles move through varying gas density is a meaningful extension of prior 2D and single-fluid studies. The qualitative mechanism is convincingly illustrated by the example trajectory in Figure 6, which shows a 10 micron grain following the meridional inflow and passing just outside the planetary envelope into the inner disk. However, the quantitative crossing fractions presented in Figures 7 and 8 rest on an assumed dust diffusivity (Schmidt number unity) and on an unspecified initial vertical particle distribution; these assumptions are not stress-tested, so the numerical percentages should be treated with caution until clarified or supplemented with sensitivity runs.","major_comments":[{"comment":"The reported 10-30% crossing fractions for 10-100 micron grains depend directly on the assumed particle diffusivity D = D_g/(1+St^2) with D_g = alpha c_s H, i.e., a Schmidt number of unity. The advective crossing mechanism shown in Figure 6 requires grains to be present at high altitude (z ~ 2H, where the meridional inflow is strong). With a more conservative Schmidt number of 3-10, the equilibrium dust scale height H_d/H ~ sqrt(alpha/[Sc(alpha+St)]) drops from near unity to roughly 0.6-0.3, and the population at 2H falls by orders of magnitude. The paper gives no sensitivity test for Sc or for the gas diffusivity, and this is load-bearing because the mixing percentages in Section 4 are used to argue for specific levels of NC-CC exchange. The authors should either add a sensitivity study (e.g., Sc = 3 and 10) or explicitly reframe the reported percentages as illustrative upper limits.","section":"Section 2.2 (Eq. 6) and Section 3.2 (Figures 7-8)"},{"comment":"The normalization of the crossing fraction is ambiguous. The sentence 'additional particles are integrated for each planet mass until 1000 total particles drift inwards of 5.2 au by 1 Myr' suggests that the number of crossing particles is fixed at 1000, which would make the reported crossing fraction depend on the stopping rule. Please specify the exact numerator and denominator used for the fractions in Figure 8, and clarify whether particles that leave the simulation bounds ('Out of Bounds' in Figure 7) are included in the denominator. Without this information, the 30% figure cannot be independently evaluated.","section":"Section 3.2, paragraph beginning 'To explore the fraction...'"},{"comment":"The initial vertical distribution of the particles is not stated. If 'evenly distributed in the 7 to 12 au region' means a uniform distribution in z up to the simulation boundary (about 3 pressure scale heights), the initial condition overpopulates high altitudes relative to a physically motivated settled distribution. Because the crossing mechanism relies on grains starting at, or diffusing to, high z, this choice can inflate the crossing efficiency. Please specify the initial z distribution and test the sensitivity to an initially settled distribution with dust scale height H_d derived from the same diffusion model.","section":"Section 3.2 (initial conditions) and Section 2.2"}],"minor_comments":[{"comment":"The random displacement uses p_x, p_y, p_z drawn from a uniform distribution on [-1,1] rather than a Gaussian. The variance is correctly normalized (2D Delta t per dimension), but the uniform distribution has bounded support and no tails; over many steps the central limit theorem applies, but the difference could matter for short integrations or near boundaries. A brief justification or reference would be helpful.","section":"Section 2.2, Eq. (3)"},{"comment":"The timestep coefficients xi = 1e-6 and zeta = 1e-5 are stated to have been tested at an order of magnitude smaller with no significant difference, which is good; however, the reported convergence test is not shown. A sentence describing the metric used for 'no significant difference' would make this statement more quantitative.","section":"Section 2.2, Eqs. (18)-(21)"},{"comment":"The vertical domain extends about 3 scale heights; particles that diffuse above this are removed. Because the advective inflow region may extend to several scale heights, the paper should comment on whether the vertical boundary is high enough to capture the full inflow region, and whether the 'Out of Bounds' loss affects the inferred crossing statistics.","section":"Section 2.1 (mesh) and Section 3.2 (boundary losses)"},{"comment":"The sentence 'This affect can also be seen in 5' contains a typo ('affect' should be 'effect') and 'in 5' should be 'in Figure 5'.","section":"Section 3.3, text near Figure 9"},{"comment":"The caption reads 'In the high viscosity case, the solids diffuse more vertically and tend to concentrate near the gap. In the high viscosity case, solids much more closely follow the gas advection...'; the second sentence should presumably read 'low viscosity case'.","section":"Figure 13 caption"},{"comment":"The citation 'Price et al. in press' in Section 2.2 does not appear in the reference list; please update to a published or arXiv reference.","section":"References"},{"comment":"The exponential term has mismatched parentheses: 'exp[(1 - sin(theta)^{-2gamma})/(2gamma h^2]' is missing a closing parenthesis. Please check the formula and ensure it matches the standard hydrostatic density profile.","section":"Appendix A, Eq. (A1)"}],"recommendation":"major_revision","confidential_remarks":"The paper is a solid and interesting contribution within the journal's scope. The qualitative mechanism is well supported by the example trajectory, and the method is a useful extension of prior work. My main concerns are (1) the lack of sensitivity testing for the Schmidt number and initial vertical distribution, which directly affects the quantitative crossing percentages that the discussion sections rely on, and (2) the ambiguous normalization of the crossing fraction in Section 3.2. These are fixable with additional runs or by reframing the quantitative claims as illustrative. I would not recommend rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear [Colleague],\n\nThe thing to know: this paper does something genuinely new—it takes the Ciesla Monte Carlo particle-tracking method into 3D, non-axisymmetric FARGO3D gas fields—and the central result, that small well-coupled grains can cross the gap of a massive embedded planet by riding the meridional inflow onto the planet, is physically plausible and consistent with earlier 3D gas-flow studies. The example trajectory in Figure 6 is strong evidence the mechanism exists. The paper also shows a secondary dust pileup in the outer disk for a roughly Jupiter-mass embryo, a useful new wrinkle for the NC/CC/CI trichotomy discussion. I'd send this to peer review; the qualitative result deserves referee time.\n\nWhere I agree with the reader's conditionality is on the numbers. The 10–30% crossing fractions in Figures 7 and 8 depend directly on the Schmidt number unity assumption in Eq. (6). If Sc is 3–10, the equilibrium dust scale height drops from ~H to ~0.3–0.6H, and the population at the altitudes where the crossing flow lives falls by an order of magnitude. The paper provides no check of Sc or of the initial vertical distribution. That makes the percentages illustrative rather than robust. The qualitative claim—that advection can carry dust across the gap—does not collapse, because it only requires that some grains get lofted, not a specific fraction.\n\nTwo smaller issues. The text says \"Eulerian approach\" in Eq. (3) but then says \"We use a Runge-Kutta 4th order method.\" That inconsistency should be fixed; it matters for reproducibility. And the conditioning in the particle re-integration step—\"additional particles are integrated ... until 1000 total particles drift inwards\"—is unclear. If the sample is selected on the outcome, the reported fractions could be biased. The paper should state exactly how many particles were integrated per case and whether crossing is a stopping criterion.\n\nThe limitations section is honest about constant gas structure, no migration, and finite particle number. The citations are appropriate. This is a solid, useful paper for the planet-formation/meteoritics crowd. It just needs the diffusion law and the particle statistics cleaned up before anyone quotes the percentages in a meteorite interpretation.\n\nI'd bring it to reading group and would cite the qualitative mechanism in my own work.","headline":"The 3D tracking is a real advance and the meridional-advection crossing mechanism holds up qualitatively, but the quoted crossing fractions lean heavily on an under-tested Schmidt-number-unity diffusion law and some integration details are sloppy.","tokens_in":26773,"tokens_out":4848,"would_cite":true,"duration_ms":44803,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A growing giant planet does not fully isolate the inner disk: gas falling onto the planet sweeps small, well-coupled dust across the planet's gap, and up to 30% of grains smaller than 100 $\\mu$m cross from the outer to inner disk within 1…","keywords":["protoplanetary disks","planet-disk interaction","gap opening","dust transport","Monte Carlo particle tracking","meridional flows","isotopic reservoirs","Jupiter formation"],"falsifier":"Rerun the particle code with the diffusion relation changed to $D = \\nu/(\\mathrm{Sc}(1+\\mathrm{St}^2))$ for $\\mathrm{Sc}=3$ and recompute the 1-Myr crossing fraction for 10 $\\mu$m grains at a 300 $M_\\oplus$ embryo; if the fraction stays near the reported ~30%, the mechanism is not tied to the unit-Schmidt-number assumption, but if it drops below a few percent, the reported mixing rates are falsified as the natural prediction of this model.","tokens_in":25694,"feed_emoji":"🪐","tokens_out":9739,"duration_ms":91172,"temperature":0.7,"pith_summary":"This paper asks whether an embedded giant planet, usually pictured as a barrier between inner and outer protoplanetary disk material, actually lets solids cross its gap. Combining 3D hydrodynamic gas simulations with Monte Carlo particle tracking, it finds that once the planet exceeds roughly the mass needed to isolate pebbles, small dust grains that are tightly coupled to the gas are swept across the gap by gas flowing onto the planet. Up to 30% of grains smaller than 100 $\\mu$m filter from the outer disk to the inner disk within 1 Myr, and 10 $\\mu$m dust can move outward as well. If correct, this means Jupiter's growth mixed, rather than strictly separated, the isotopic reservoirs recorded in meteorites, with the amount of mixing set by grain size, turbulent viscosity, and embryo mass.","feed_headline":"Up to 30% of fine dust crosses a giant planet's gap","feed_subtitle":"3D simulations show sub-100-micron grains filter past the growing Jupiter in both directions, reshaping the meteorite record.","key_machinery":"The load-bearing object is the planet's 3D meridional gas circulation: outward along the midplane, inward at altitude. The particle integrator carries the argument: each grain is moved by stellar and planetary gravity, gas drag with a locally evaluated Stokes number, and a stochastic diffusion step whose amplitude comes from $D=\\nu/(1+\\mathrm{St}^2)$ with $\\nu=\\alpha c_s H$, i.e. a unit Schmidt number. Because small grains diffuse to altitudes where the flow is inward, and because their Stokes number rises as gas density drops inside the gap, the model predicts which sizes remain coupled long enough to be funneled past the planet. It also yields radial residence times, showing where grains pile up at one or more outer pressure bumps.","core_discovery":"The central discovery is that the 3D structure of the gas flow around an embedded giant planet reverses the usual picture of gap physics. Near the midplane the gas flows outward from the planet, but at higher altitudes it flows inward onto the planet. Small, well-coupled grains that diffuse vertically into these upper layers are entrained in the inward advective flow, pass within a few hundred Jupiter radii of the planet yet outside its accretion envelope, and can be deposited on orbits interior to the planet. For planets above the pebble isolation mass, conventionally placed near $30\\,M_\\oplus$ in the fiducial disk and efficient by $100\\,M_\\oplus$, this lets up to 30% of sub-100-$\\mu$m particles cross outward to inward within 1 Myr while millimeter- and centimeter-sized pebbles remain trapped at the pressure bump. The same meridional circulation carries 10 $\\mu$m dust from the inner disk outward, and in low-viscosity disks the massive planet creates multiple outer pressure bumps that may support a third isotopic reservoir.","pith_inferences":["If the real dust diffusivity is lower than gas momentum diffusivity (Schmidt number greater than one), the reported 10–30% crossing fractions are upper limits; rerunning the same code with $\\mathrm{Sc}=2$\\u2013$5$ would bracket the mixing efficiency.","The advection-funneling mechanism is not Jupiter-specific and should operate around any sufficiently massive planet in a viscous disk, meaning isotopic reservoir separation in other planetary systems may also be leaky.","A testable consequence of the funneling effect is a spatial correlation between isotopic anomalies and thermal-processing signatures in small CAIs that crossed the gap, which could be searched for in individual inclusions.","Planet migration would shift the balance: an inward-migrating planet sees a relative outward gas flow, which should suppress inward crossing and enhance outward transport of small dust; adding migration to the particle tracking is the natural next test."],"forward_implications":["Above the pebble isolation mass a giant planet acts as a size-selective filter rather than a closed barrier: grains smaller than about 100 $\\mu$m leak across the gap in both directions while larger pebbles stay trapped at the pressure bump.","The extent of mixing changes as Jupiter grows: large grains drift inward freely before the isolation mass, sub-100-$\\mu$m grains cross efficiently once the embryo reaches roughly $100\\,M_\\oplus$, and at the highest masses only 10 $\\mu$m dust remains coupled enough to cross.","In low-viscosity disks a single massive planet can create multiple outer pressure bumps, so dust may pile up in several rings sourced from different parts of the disk, plausibly preserving a third isotopic reservoir.","Grains that cross are funneled within a few hundred Jupiter radii of the planet, so late-arriving inner-disk solids would have been thermally processed near the accreting protoplanet, potentially altering their volatile and isotopic content.","In exoplanetary systems, an outer giant planet does not necessarily cut off the supply of small dust to the inner disk, so inner super-Earth formation could still receive feedstock after the outer planet opens a gap."],"supporting_citations":[{"why":"Supplies the Monte Carlo particle-tracking method, including the diffusion-biased effective velocity, used to follow individual dust grains through the gas fields.","marker":"Ciesla (2010, 2011)"},{"why":"Gives the relation $D = D_g/(1+\\mathrm{St}^2)$ relating particle diffusivity to gas diffusivity and Stokes number, which controls how far grains scatter vertically.","marker":"Youdin & Lithwick (2007)"},{"why":"Earlier 2D simulations showing small solids filtering past an embedded giant planet; this paper's 3D advection crossing is contrasted with that filtering.","marker":"Weber et al. (2018)"},{"why":"3D simulations of meridional gas flows and small-dust accretion onto embedded planets, the flow pattern this paper invokes to carry grains across the gap.","marker":"Szulágyi et al. (2022)"},{"why":"Defines the pebble isolation mass and the outer pressure bump that halts pebble drift, the mass threshold around which crossing behavior changes.","marker":"Lambrechts et al. (2014)"},{"why":"The NC-CC isotopic dichotomy attributed to a Jupiter-opened gap is the meteoritic constraint the model is compared against.","marker":"Kruijer et al. (2017)"},{"why":"Documents the size and abundance difference of CAIs between NC and CC chondrites, the size-selective pattern the model can explain.","marker":"Dunham et al. (2023)"}],"fun_headline_variants":["3D gas flow lets fine dust slip past giant planets both ways","Up to 30% of fine dust sneaks across giant planet's gap","Growing Jupiter's 3D flow mixes inner and outer disk dust","Bidirectional dust transport: how small grains cross planet gaps","Planet's 3D wake drives dust across gap in both directions"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The quantitative crossing fractions rest on the assumption that dust diffuses in the gas exactly as fast as momentum does (a unit Schmidt number) and that the diffusion is an isotropic random walk; if diffusion is weaker or anisotropic, fewer small grains would be lofted into the inward-flowing layers and the 10–30% crossing fractions would shrink.","fun_headline_variants_meta":{"raw":{"variants":["3D gas flow lets fine dust slip past giant planets both ways","Up to 30% of fine dust sneaks across giant planet's gap","Growing Jupiter's 3D flow mixes inner and outer disk dust","Bidirectional dust transport: how small grains cross planet gaps","Planet's 3D wake drives dust across gap in both directions"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000683,"raw_usage":{"total_tokens":3113,"prompt_tokens":973,"completion_tokens":2140,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":589,"completion_tokens_details":{"reasoning_tokens":2048}},"tokens_in":589,"tokens_out":2140,"duration_ms":13095,"temperature":1.0,"reasoning_tokens":2048,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T20:39:05.334659+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Rerun the particle code with the diffusion relation changed to $D = \\nu/(\\mathrm{Sc}(1+\\mathrm{St}^2))$ for $\\mathrm{Sc}=3$ and recompute the 1-Myr crossing fraction for 10 $\\mu$m grains at a 300 $M_\\oplus$ embryo; if the fraction stays near the reported ~30%, the mechanism is not tied to the unit-Schmidt-number assumption, but if it drops below a few percent, the reported mixing rates are falsified as the natural prediction of this model.","supporting_citations":[],"review_version":1}